Tethered IL-15 augments antitumor activity and PNAS PLUS promotes a stem-cell memory subset in tumor-specific T cells

Lenka V. Hurtona,b, Harjeet Singha, Amer M. Najjarc, Kirsten C. Switzera, Tiejuan Mia, Sourindra Maitia, Simon Olivaresa, Brian Rabinovicha, Helen Hulsa,d, Marie-Andrée Forgeta,e, Vrushali Datarc, Partow Kebriaeif, Dean A. Leea, Richard E. Champlinf, and Laurence J. N. Coopera,g,1

aDivision of Pediatrics, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030; bThe University of Texas Graduate School of Biomedical Sciences, Houston, TX 77030; cCancer Systems Imaging, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030; dIntrexon, Germantown, MD 20876; eMelanoma Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030; fStem Cell Transplantation and Cellular Therapy, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030; and gZIOPHARM Oncology, Inc., Boston, MA 02129

Edited by Warren J. Leonard, National Heart, Lung and Blood Institute, Bethesda, MD, and approved October 20, 2016 (received for review June 28, 2016) + + + Adoptive immunotherapy retargeting T cells to CD19 via a chimeric CD8 CD45RA CCR7 TSCM-like cells in the infusion product antigen receptor (CAR) is an investigational treatment capable of correlated with expansion of CD19-specific CAR T cells (11). inducing complete tumor regression of B-cell malignancies when Because TSCM represent only a small percentage (2–3%) of pe- there is sustained survival of infused cells. T-memory stem cells ripheral blood mononuclear cells (PBMCs) (8), strategies to (T ) retain superior potential for long-lived persistence, but chal- SCM manufacture TSCM suitable for human applications are essential lenges exist in manufacturing this T-cell subset because they are rare and under development (9). among circulating lymphocytes. We report a clinically relevant ap- + Endogenous and administered T cells receive prosurvival sig- proach to generating CAR T cells with preserved TSCM potential nals through the common γ-chain, such as those using the Sleeping Beauty platform. Because IL-15 is fundamental signals mediated by IL-2 and IL-15, independent of native or in- to T-cell memory, we incorporated its costimulatory properties by troduced immunoreceptors. ACT trials have provided exogenous coexpressing CAR with a membrane-bound chimeric IL-15 (mbIL15). INFLAMMATION IL-2, which causes dose-limiting toxicities at high doses (12), IMMUNOLOGY AND The mbIL15-CAR T cells signaled through signal transducer and whereas low doses may favor peripheral tolerance and regulatory activator of transcription 5 to yield improved T-cell persistence independent of CAR signaling, without apparent autonomous T-cell production (13). Moreover, IL-2 signaling drives effector T-cell growth or transformation, and achieved potent rejection of proliferation, promoting terminal differentiation and senescence + neg + + (14). Conversely, IL-15 is a prosurvival cytokine that is required for CD19 . Long-lived T cells were CD45RO CCR7 CD95 , + homeostatic maintenance of long-lived CD8 memory T cells (15), phenotypically most similar to TSCM, and possessed a memory-like transcriptional profile. Overall, these results demonstrate that inhibits activation-induced cell death (AICD) (16), enhances in CAR+ T cells can develop long-term persistence with a memory vivo antitumor activity (17), and reverses T-cell anergy (18). High stem-cell phenotype sustained by signaling through mbIL15. This IL15 expression in the tumor microenvironment correlates with + observation warrants evaluation in clinical trials. elevated infiltration of CD3 T cells, correlating with improved

adoptive immunotherapy | CAR | IL-15 | T-memory stem cell | Significance T-cell persistence We describe an approach based on cytokine therapeutics to himeric antigen receptors (CARs) that redirect T-cell spec- enhance the persistence and effectiveness of T-cell–based im- Cificity to desired tumor-associated antigens (TAAs) (1) are munotherapies using chimeric antigen receptors (CARs). This engineered to activate T cells for survival, serial killing, and cytokine strategy is effective without the use of high-dose exogenous production only upon contacting TAA (2). Adoptive transfer of CAR cytokines that are typically associated with toxicities. Moreover, T cells can achieve durable complete responses in some patients; we report that the persistence of the least differentiated mem- successful outcomes are associated with engraftment and long-term ory , the T-memory stem cell, was promoted by signaling persistence of CAR T cells (3). Long-term immunosurveillance by induced by a membrane-bound chimeric IL-15 cytokine-fusion persisting CAR T cells is likely key to achieving durable responses in molecule. These findings may contribute to improving the safety adoptive cell therapy (ACT). Memory T-cell subsets appear to exist and therapeutic efficacy of CAR-based immunotherapies of pa- along a gradient of differentiation characterized by reciprocal po- tients with advanced cancer. tentials for longevity and effector function (4). Indeed, adoptively + transferred effector CD8 T cells derived from central memory Author contributions: L.V.H., H.S., A.M.N., H.H., M.-A.F., and L.J.N.C. designed research; L.V.H., A.M.N., K.C.S., T.M., S.M., and V.D. performed research; S.O. and B.R. contributed (TCM)ornaive(TN) T-cell subsets in murine and nonhuman primate new reagents/analytic tools; L.V.H. analyzed data; and L.V.H., P.K., D.A.L., R.E.C., and L.J.N.C. models demonstrated increased therapeutic potential. Thus, T-cell wrote the paper. subsets corresponding to an immature state of differentiation are Conflict of interest statement: Some of the technology described was advanced through appealing for their potential to provide superior clinical utility (5, 6). research conducted at the M. D. Anderson Cancer Center by L.J.N.C. The M. D. Anderson T-memory stem cells (TSCM), so far the least differentiated Cancer Center, L.J.N.C., L.V.H., K.C.S., H.S., T.M., S.M., S.O., B.R., M.-A.F., H.H., A.M.N., R.E.C., memory T-cell subset identified, can be generated under specific and D.A.L. all have or had a financial interest in ZIOPHARM Oncology, Inc., and Intrexon Corporation. On May 7, 2015, L.J.N.C. was appointed as the Chief Executive Officer at ex vivo culture conditions (e.g., IL-7, IL-15, or small molecules ZIOPHARM. L.J.N.C. is now a Visiting Scientist at the M. D. Anderson Cancer Center. A patent targeting metabolic or developmental pathways) (7–9). This application based on this manuscript has been filed. memory subset possesses the highest self-renewal capacity and This article is a PNAS Direct Submission. therapeutic potential. Due to superior persistence in the absence Freely available online through the PNAS open access option. of antigen-driven stimulation, TSCM are suggested to be the 1To whom correspondence should be addressed. Email: [email protected]. primary precursors of T-cell memory once antigen is cleared in This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. an immune response (10). Furthermore, only the frequency of 1073/pnas.1610544113/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1610544113 PNAS Early Edition | 1of10 Downloaded by guest on October 1, 2021 survival of patients with colorectal cancer (19). Moreover, IL-15 is for treating and preventing relapse when antigen burden is high required for the generation of innate-like T cells that participate in but also for treating malignant diseases with low or sequestered immunosurveillance and impede tumor growth (20). tumor burdens. IL-15 is recognized as an immunotherapeutic agent for cancer treatment (21) and is being coinfused with natural killer (NK) or Results T cells. Monomeric IL-15 is a small unstable with a short AaPC-Based Platform Generates Numerous T Cells Stably Coexpressing serum t1/2, necessitating supraphysiological dosing to attain in CAR and MbIL15 with Enhanced Signaling via Phosphorylated Signal vivo responses (22). However, physiological transpresentation of Transducer and Activator of Transcription 5. Second-generation CARs IL-15 uses the IL-15 receptor alpha chain (IL-15Rα) (23, 24). mediate signal 1 (via CD3-ζ) and signal 2 (e.g., via CD28) for T-cell Recombinant soluble IL-15 bound to a recombinant soluble IL- activation. We designed mbIL15 for cell-surface expression to 15Rα (referred to as IL-15 complex) augments T-cell antitumor provide signal 3 (cytokine stimulation), with and without CAR- activity (22), likely due to the qualitatively different signaling mediated signaling. To generate mbIL15, the full-length native IL- achieved upon mimicking transpresentation (25). Thus, IL-15 15 peptide was fused to the full-length IL-15Rα sequence via a complex and ALT-803 (a fusion protein of IL-15 superagonist flexible linker (Fig. 1A), mimicking presentation of IL-15 in context and IL-15Rα ) have enhanced the antitumor effect with IL-15Rα (30, 31). T cells from PBMCs were coelectroporated of responding T cells (26, 27), leading to testing in clinical trials. with DNA plasmids coding a CD19-specific CAR transposon To improve the persistence and potential for memory of in- (designated CD19RCD28), with or without the mbIL15 transposon fused T cells, we sought to harness IL-15 transpresentation by and with the SB-11 transposase. These cells were propagated engineering a recombinant tethered variant of IL-15 coexpressed with small modifications from previously described methods adapted with a second-generation CAR using the Sleeping Beauty (SB) for human application (28) (SI Appendix,Fig.S1). CAR T cells system (28). The SB system is the most clinically advanced generated with the SB platform were recursively stimulated using nonviral approach to therapy, with recent trials demon- K562-derived AaPCs ex vivo to propagate genetically modified T strating promising data related to survival of infused CD19- cells in an antigen-dependent fashion to yield a high frequency of specific CAR T cells and recipients with B-cell malignancies cells expressing CAR in the final product. To expand out T cells (29). Signaling from membrane-bound IL-15 (mbIL15) gener- selectively coexpressing both mbIL15 and CAR after electroporation ated CD19-specific CAR T cells with long-term persistence and (Fig. 1B), we added recombinant soluble IL-21 along with AaPCs superior in vivo antitumor activity. The mbIL15-CAR T cells, during ex vivo coculture to capitalize on the synergistic effect of despite being recursively cocultured with γ-irradiated activating combining IL-21 with IL-15 upon T-cell proliferation (32). After four and propagating cells (AaPCs), exhibited an immature state of 8- to 10-d cycles of γ-irradiated K562-derived AaPCs (clone 9), CAR differentiation upon antigen removal, leading to sustained in expression on the CAR and mbIL15-CAR T cells was not signifi- vitro and in vivo persistence without aberrant T-cell proliferation. cantly different [94% ± 6% (mean ± SD) and 94% ± 7%, re- Indeed, combining expression of mbIL15 with CAR generated T cells spectively; P = 0.9748, n = 10; Fig. 1B). Electrotransfer of mbIL15 + + that retained memory potential with a CD45ROnegCCR7 CD95 and CAR from two separate plasmids resulted in coexpression of TSCM-like phenotype. Because mbIL15-CAR T-cell survival appears mbIL15 and CAR at 51% ± 14% (mean ± SD; n = 10) on the independent of antigen, this immunotherapy may be of use not only antigen-stimulated cells, with minimal expression of mbIL15 alone

A D 10 12 mbIL15-CAR IgE Signal 11 CAR Peptide Ser-Gly 10 α 10 IL-15 Linker IL-15R Flag 10 ns 5’ 3’ 10 9

1266 bp T cell count 8 + 10 B CAR mbIL15-CAR 10 7 73 1 60 11 10 6 Fig. 1. T cells stably coexpressing mbIL15 and CAR 26 0 29 0 display enhanced costimulation via pSTAT5. (A) Dia- 10 5 24-Hours Post- gram of the mbIL15 coding insert used within the 4 Electroporation Inferred CAR 10 SB DNA expression vector [IL15-IL15Ra-Flag(CoOP)/ 01234 pSBSO]. (B) Transient (24-h postelectroporation) and Stim number stable [stimulation 4 (Stim 4)]) expression of mbIL15 CAR 98 1 50 49 10 10 E 100 mbIL15-CAR and CAR on SB-modified T cells. A representative Stim 4 80 CAR & soluble IL-15 flow plot from one of 10 normal donors is shown. CAR (no cytokine) Quadrant statistics display the percentage of gated 60 + + Isotype cells. (C) Frequencies of mbIL15 and/or CAR subsets 40 of CAR and mbIL15-CAR T cells. Data are mean ± SD % of Max 20 mbIL15 (n = 10). (D) Inferred numeric expansion of CAR and 0 γ 0101 102 103 104 mbIL15-CAR T cells upon serial coculture with -irradi- mbIL15-CARmbIL15-CAR ated AaPC (“Stim number”) and cytokines. Data are C pSTAT5 100 CARCAR expressed as mean ± SD (n = 6), two-way repeated

%) F mbIL15-CAR measures ANOVA (Bonferroni posttest). (E) Functionality CAR 75 100 of mbIL15 was validated using phosflow of pSTAT5 expression in CAR and mbIL15-CAR T cells under serum- 80 50 and cytokine-starved conditions. Positive control CAR T 60 cells were treated with 100 ng/mL IL-15 for 1 h before

25 sion (%) 40 fixation. A representative experiment is shown for one * of six donors. Max, maximum. (F) Phenotype of CAR and 0 pres 20 Subset Frequency ( mbIL15-CAR T cells was assessed by flow cytometry for + + Ex neg neg 0 T-cell markers and surface markers associated with + + neg neg 4 A α 7 memory and differentiation state after Stim 4 expansion. mbIL15 mbIL15 mbIL15CAR mbIL15CAR CD3 CD CD8 CAR CAR L-7R CCR CD27 CD28 Horizontal lines indicate mean values, and dots represent Subsets I CD62L CD45RCD45RO individual donors. *P = 0.047 (n > 4), paired t test.

2of10 | www.pnas.org/cgi/doi/10.1073/pnas.1610544113 Hurton et al. Downloaded by guest on October 1, 2021 PNAS PLUS ABAWD-mbIL15-CAR mbIL15-CAR 225 # # Isotype 25 100 80 mbIL15-CAR 20 60 10 # 40 0 % of Max 20 # # 0 -10 # 0101 102 103 104 0101 102 103 104 0101 102 103 104

Fold Change TCF-1 T-BET BLIMP-1 -20 -25 ## D E -225 AWD- 100 mbIL15-CAR mbIL15-CAR 1 4

D-mbIL15-CAR relative to 80 LY

MH 93 4 452 11B IL7R LEF1 IFNG SELL BCL6 TCF7 BCL2 CD28 PRF1 CCR7 GN 30 044 IL2RA 60 GZMB GZ AW TBX21 CTLA CD FOXP1 BACH2 PRDM EOMES 40

Transcriptional Cytotoxicity Activation % of Max 20

Regulators or Trafficking CD45RO 0 0101 102 103 104 CD45RA C 100 CD45RA 80 F 60 AWD-mbIL15-CAR mbIL15-CAR 40 ** * ** *** ** * 20 100 0

1 2 3 4 1 2 3 4 1 2 3 4 )

01010 10 10 01010 10 10 01010 10 10 80 INFLAMMATION IL-2Rα IL-7Rα CCR7 IMMUNOLOGY AND 60 100 on (% * * % of Max 80 ssi 40 60 pre 20 40 Ex 0 20 α α 7 0 1 2 3 4 1 2 3 4 1 2 3 4 01010 10 10 01010 10 10 01010 10 10 CD27 CD28 IL-2R IL-7R CD62L CCR CD27 CD28 CD62L CD45RA CD45RO

Fig. 2. Prolonged in vitro antigen-independent persistence of mbIL15-CAR T cells creates a molecular profile and phenotype associated with an immature dif- ferentiation state. After four stimulation cycles with γ-irradiated AaPCs (35 total days of culture), mbIL15-CAR T cells were placed under antigen and exogenous cytokine withdrawal conditions for 65 d (100 total days of culture after electroporation). T cells persisting in culture for >65 d are referred to as “AWD-T cells.” The CD4/CD8 ratio of starting cell populations was 1:39.9 ± 58.6 (mean ± SD). (A) Comparison of the top significantly differentially expressed (twofold cutoff, P < 0.01, false discovery rate q value <0.05). Classified by function, these genes are associated with immature differentiation (blue bars) or more differentiated (red bars) effector T-cell characteristics. Bars represent mean fold change of AWD-mbIL15-CAR T cells relative to mbIL15-CAR T cells (n = 3). The shaded gray area denotes less than twofold change. #, verification of protein expression by flow cytometry. (B) Representative histograms of transcription factor expression associated with differentiation status. (C) Representative histograms of surface-expressed molecules associated with differentiation status. (D) Representative flow plots showing coexpression of CD45RA and CD45RO. Quadrant frequencies are displayed in the top right corner of the plots. (E) Representative histogram of CD45RA expression + on CCR7 gated cells. (F) Frequencies of respective markers from C and D. Data are plotted as box and whiskers, with the horizontal bar denoting the median. Data in B–F are representative of greater than five donors. *P < 0.05, **P < 0.01, ***P < 0.001; paired t test.

(Fig. 1 B and C). Ex vivo expansion of mbIL15-CAR T cells yielded MbIL15-CAR T Cells Sustain Functional IL-15 Signaling and Antigen- comparable inferred cell numbers to control CAR T cells (Fig. 1D). Specific Response in Vitro Even in the Long-Term Absence of CAR- Western blotting confirmed mbIL15 protein expression with the Mediated Activation. Activation via IL-15 should sustain T-cell anticipated molecular weight as a fusion protein (SI Appendix,Fig. persistence, which was initially assessed in vitro by withdrawing S2A). Validation of mbIL15 function assessed phosphorylation levels CD19 antigen and exogenous cytokine. Propagated mbIL15- of signal transducer and activator of transcription 5 (STAT5). The CAR T cells underwent prolonged culture without restimulation mbIL15-CAR T cells cultured without exogenous cytokines and or exogenous cytokines, whereas control CAR T cells received CAR T cells exposed to exogenous IL-15 exhibited elevated pSTAT5 either IL-15 complex or no exogenous cytokine (SI Appendix, expression, relative to CAR T cells without cytokine (Fig. 1E). Other Fig. S3A). As anticipated, CAR T cells without cytokine sup- than low expression of IL-7Rα, mbIL15 expression did not alter the plementation did not persist (SI Appendix,Fig.S3B). CAR T cells phenotype of ex vivo expanded CAR T cells (Fig. 1F). The mbIL15- receiving IL-15 complex initially expanded, and then maintained CAR product produced by coculture with AaPCs and IL-21 was cell numbers. In contrast, mbIL15-CAR T cells did not expand, but + + predominantly CD8 CD45RO T cells expressing moderate levels of persisted without unrestricted growth, and cell viability of the CD62L and CD28 (Fig. 1F). Antigen-specific cytolysis and IFN-γ groups was similar at day 65 (P > 0.05, Student’s t test; SI Appendix, production remained intact and comparable between CAR and Fig. S3 B and C). The mbIL15-CAR T cells maintained a constant mbIL15-CAR T cells (SI Appendix,Fig.S2D and E). Thus, clinically cell number and did not grow in an unrestricted or autonomous large numbers of functional mbIL15-CAR T cells possessing elevated manner. T cells persisting at and beyond 65 d under antigen- pSTAT5 can be generated. withdrawal (AWD) conditions maintained CAR expression

Hurton et al. PNAS Early Edition | 3of10 Downloaded by guest on October 1, 2021 A AWD-mbIL15-CAR B 100 mbIL15-CAR ** **** 80 % of Max

60 ns ns 0101 102 103 104 40 CD95 Isotype Expression (%) 20 CD45ROnegCCR7neg Fig. 3. Emergence of a TSCM-like subset after pro- CD45RO+CCR7neg longed in vitro antigen-independent persistence of 0 CD45RO+CCR7+ mbIL15-CAR T cells. After four stimulation cycles with γ-irradiated AaPCs (35 total days of culture), mbIL15- CD45ROnegCCR7+ CAR T cells were placed under antigen and exoge- nous cytokine withdrawal conditions for 65 d (100 **** total days of culture after electroporation). T cells 7 10 C mbIL15-CAR > ns persisting in culture for 65 d are referred to as *** *** 100 “AWD-T cells.” The CD4/CD8 ratio of starting 80 cell populations was 1:9.6 ± 10.0 (mean ± SD). 4.5% 60 (A) Memory subset (based on CD45RO and CCR7) 10 6 composition (Top) and cell counts (Bottom). Data are

ber 40 plotted as box and whiskers, with the horizontal bar

% of Max 20 denoting the median, and are representative of greater than five donors. **P < 0.01, ***P < 0.001, 0 ****P < 0.0001, two-way ANOVA (Bonferroni’s ll Num 5 10 AWD-mbIL15-CAR posttest). ns, not significant. (B) At the end time Ce 100 point, AWD-mbIL15-CAR T cells were assessed for + 80 CD95 expression to dissect the CD45ROnegCCR7 33.2% population further and delineate T from T sub- 60 SCM N 10 4 sets. Representative histogram of one of four donors 40 showing CD95 expression on cells gated on memory CD45ROneg CD45RO+ CD45RO+ CD45ROneg + + neg neg % of Max 20 subsets from A. The vertical bar in the plot denotes a CCR7 CCR7 CCR7 CCR7 reference line based on isotype staining. (C) Repre- 00101 102 103 104 sentative histograms of IL-2 intracellular staining af- Differentiation ter stimulation with lymphocyte activation cocktail IL-2 (black line) or no stimulation (gray) (n = 6).

(SI Appendix,Fig.S3D and E). Although low surface ex- pendix, Fig. S2C), indicating no initial overt molecular survival pression of mbIL15 was observed on AWD-mbIL15-CAR T advantage for mbIL15-CAR T cells. cells compared with day 0, intracellular levels were detected Differences emerged when comparing persisting AWD- (72% ± 23%) (SI Appendix,Fig.S3D and E). Moreover, we mbIL15-CAR cells with mbIL15-CAR T cells, where 98 genes observed sustained expression of pSTAT5 and Bcl-2, a pro- were significantly differentially expressed (SI Appendix, Fig. S4). survival molecule downstream of STAT5, confirming contin- Transcripts encoding key regulators of effector differentiation ued signaling pathway activation (SI Appendix,Fig.S3F). Both and related cytotoxic molecules showed reduced expression in AWD-CAR (cytokine supplemented) and AWD-mbIL15-CAR AWD-mbIL15-CAR T cells: TBX21 (encoding T-BET), PRDM1 T cells remained antigen-responsive, producing comparable (encoding BLIMP-1), IL2RA, IFNG, PRF1, GNLY, GZMH, + IFN-γ levels upon coculture with CD19 targets (P > 0.05; SI GZMB, CD11B,andCTLA4. Inversely, a promemory tran- scriptional profile indicating an immature state of differentia- Appendix,Fig.S3G). Thus, long-term persisting mbIL15-CAR T tion was present in AWD-mbIL15-CAR T cells: TCF7 (encoding cells retain functional mbIL15 and effector responses in the TCF-1), EOMES, BCL2, BACH2, LEF1, BCL6, FOXP1, CCR7, absence of CD19 TAA. and IL7R (8, 33) (Fig. 2A and SI Appendix, Fig. S4). Some of these transcription factors function in pairs, with graded re- Long-Term in Vitro Antigen-Independent Persistence of MbIL15-CAR T ciprocal activity controlling effector and memory T-cell differ- Cells Exhibits a Molecular Profile and Phenotype Associated with + entiation. The ratio of TBET and EOMES is one such pairing, T . Because IL-15 plays a crucial role in CD8 memory T-cell SCM with the ratio lowest in memory T-cell stages (33). We observed homeostasis (15), we postulated that mbIL15 could support long- that mbIL15-CAR and AWD-mbIL15-CAR T cells had altered lived TSCM in vitro in the absence of immunoreceptor signaling. To mRNA ratios of TBET and EOMES (20:1–94:1 and 2:1–9:1 assess whether mbIL15 conferred a survival advantage via molec- TBET/EOMES, respectively), with AWD-mbIL15-CAR T cells ular programming, we compared the mRNA levels of 479 lym- having the low ratio associated with memory phenotypes. phocyte-specific genes between cultured CAR and mbIL15-CAR T The protein expression of specific transcription factors and cells, both products of four AaPC stimulations, using a digital gene surface markers associated with the status of T-cell differentia- expression analysis with a lymphocyte-specific probe set (SI Ap- tion corroborated the molecular profile of a less differentiated pendix,TableS1). Surprisingly, we found only four differentially state for AWD-mbIL15-CAR T cells, with increased expression expressed genes (SI Appendix,TableS2), with the difference in of cell-surface markers associated with memory (IL-7Rα, CD27, IL7RA expression corroborated by cell-surface phenotyping (SI Ap- and CCR7; Fig. 2 B, C, and F). CD62L and CD28 expression on

4of10 | www.pnas.org/cgi/doi/10.1073/pnas.1610544113 Hurton et al. Downloaded by guest on October 1, 2021 PNAS PLUS A CAR mbIL15-CAR B CAR mbIL15-CAR 1.0 0.8 0.01 0.14 ** Day 0.6

0 Cells (%) 0.4 + Bone

Marrow 0.2

CD3 0.0 Day 6 1.5 ** 0.02 0.87 1.0 Cells (%) SSC Day + 0.5 14 Spleen

CD3 0.0

ns Day 20 40 0.00 6.35 15 Fig. 4. Long-term in vivo persistence of mbIL15-CAR neg + 10 T cells facilitates the emergence of a CD45RO CCR7

Cells (%) 7 + + subset. NSG mice were injected i.v. with 2 × 10 ffLuc Day Blood 5

47 Peripheral CAR T cells with or without coexpression of mbIL15. CD3 0 The infused T-cell number was based on CAR T cells. CD3 p/s/cm2/sr The CD4/CD8 ratio of injected cells was CAR = 1:12 3 CAR 500 (x10 ) 50 and mbIL15-CAR = 1:161. sr, steradian. (A) Longitudinal mbIL15-CAR bioluminescent imaging (BLI) was performed to monitor T-cell persistence from days 0–47. Images represent pho- C D ** ton flux from T-cell–derived ffLuc activity. (B, Left)Rep- 100 *** AWD-mbIL15-CAR resentative flow plots show frequencies of CD3 in the *** * 80 Peripheral Blood BM, spleen, and peripheral blood from tissues and blood Spleen harvested on day 48. (B, Right) Plotted frequencies of + 60 Bone Marrow CD3 cells in tissues and blood. Data are represented as 1 2 3 4 01010 10 10 mean ± SD (n = 5) and are representative of two ex- 40 CD45RO periments. *P < 0.05, **P < 0.01; unpaired t test. (C) INFLAMMATION IMMUNOLOGY AND pression (%) 20 Peripheral blood, spleen, and BM of mice were analyzed Ex after 48 d to assess the differentiation of adoptively 0 transferred T cells persisting long term in the absence of

Relative Cell Number neg + + neg CD45RO CD45RO CD45RO CD45RO CAR activation. Representative histograms of CD45RO CCR7+ CCR7+ CCR7neg CCR7neg (Upper) and CCR7 (Lower) expression. (D) Memory sub- 1 2 3 4 01010 10 10 Differentiation set composition of the in vivo T cells in reference to in CCR7 vitro AWD-mbIL15-CAR T cells. Data are plotted as box Peripheral Blood and whiskers showing the median (horizontal bar) and Spleen mean (cross). *P < 0.05, **P < 0.01, ***P < 0.001, two- Bone Marrow way ANOVA (Tukey’s posttest).

AWD-mbIL15-CAR T cells was diminished compared with To determine whether the promotion of TSCM was due to mbIL15-CAR T cells (Fig. 2 C and F). AWD-mbIL15-CAR T mbIL15, we conducted a modification of the withdrawal assay cells highly expressed CD45RA, but the majority of cells were comparing AWD-mbIL15-CAR T cells with AWD-CAR T cells + + CD45RA CD45RO (Fig. 2 D and F). However, compared with that had IL-2 or IL-15 complex support during antigen withdrawal mbIL15-CAR T cells, AWD-mbIL15-CAR T cells had a signif- (SI Appendix,Fig.S5A). CD45RA expression was uniformly high + icant population of CD45RA CD45ROneg cells [31% ± 17%, among all treatment groups, whereas CCR7 expression was only (mean ± SD); P = 0.0233, Student’s t test; Fig. 2D] with nearly all high on AWD-mbIL15-CAR T cells (69% ± 23%), which also had + CCR7 T cells coexpressing CD45RA [89% ± 10% (mean ± significantly reduced Annexin V levels compared with control + SD); Fig. 2E]. With an observed high coexpression of CD45RA CCR7 AWD-CAR T cells cultured with soluble IL-2 or IL-15 and CD45RO, we defined T-cell memory subsets by CD45RO complex (P < 0.05; SI Appendix,Fig.S5B–E). Thus, mbIL15 + and CCR7 expression (34, 35). Thus, it appears that ex vivo nu- supports the survival of CD45ROnegCCR7 T cells, thereby sup- merical expansion (with AaPCs and IL-21) generates pre- porting the persistence and/or emergence of TSCM in our system. dominantly differentiated effector memory mbIL15-CAR T cells + with a CD45RO CCR7neg phenotype (Fig. 3A). Upon the MbIL15-CAR T Cells Persisting in Vivo in the Absence of Antigen neg + withdrawal of antigen and exogenous cytokine, mbIL15 sup- Demonstrate the Emergence of a CD45RO CCR7 TSCM Subset. To ported two T-cell memory subsets. Differentiated CD45ROnegCCR7neg assess if mbIL15-CAR T cells maintain prolonged persistence in AWD-mbIL15-CAR T-cell numbers were maintained, whereas vivo without CAR-mediated signaling, genetically modified T + ± the least differentiated CD45ROnegCCR7 AWD-mbIL15-CAR cells [mbIL15 -CAR-firefly luciferase (ffLuc)] were adminis- T cells increased as much as eightfold, and the remaining two tered to mice without tumor or exogenous cytokine support and memory subsets significantly declined (Fig. 3A). All four memory monitored for 48–50 d. Control CAR T cells did not persist, subsets observed in AWD-mbIL15-CAR T cells coexpressed whereas mbIL15-CAR T cells persisted without apparent au- + CD95 (Fig. 3B), thus delineating the CD45ROnegCCR7 pop- tonomous growth and were detected primarily in the bone ulation as a TSCM subset. The gain of this immature T-cell subset marrow (BM) and spleen (Fig. 4 A and B and SI Appendix, Fig. corresponded with improved capacity for IL-2 production (Fig. S6A). Thus, it appears that mbIL15 provides sufficient support to 3C), because 2–8% of mbIL15-CAR and 25–46% of AWD- sustain T-cell persistence in a CAR-independent manner. mbIL15-CAR T cells produced IL-2 in response to stimulation, The phenotype and ability of mbIL15-CAR T cells to redirect a cardinal feature of memory cells (35). These data support specificity to CD19 were evaluated 48 d after adoptive transfer -/- our contention that mbIL15-CAR T cells contain and/or into nonobese diabetic/severe combined immunodeficiency/gc produce TSCM. (NSG) mice without tumor or exogenous cytokine support.

Hurton et al. PNAS Early Edition | 5of10 Downloaded by guest on October 1, 2021 A mbIL15-CARNo T Cell C Tumor Only CAR mbIL15-CAR 10 8 CAR No T Cell 96.2% 87.7% 0.1% 7

/sr) 10 2 Bone

Marrow Fig. 5. MbIL15-CAR T cells prevent leukemia en- 10 6 ** 0.0% 0.0% 2.8% graftment and persist after tumor clearance. NSG 7 +

(p/s/cm mice were infused i.v. with 2 × 10 ffluc CAR T cells 10 5 with or without coexpression of mbIL15. Infused 83.6% 73.5% 0.4% +

Total body T cell flux 5152535 T-cell number was based on CAR T cells. Six days later, × 5 + 10 7 1 10 rLuc NALM-6 was i.v. injected The CD4/CD8 CD19 Spleen ratio of injected cells was CAR = 1:12 and mbIL15- 0.0% 0.0% 0.0% 6 CAR = 1:161. (A) Longitudinal plotting of T-cell flux

/sr) 10 2 nfemur derived from regions of interest encompassing the *** body (Upper) and the femurs (Lower). Background 10 5 23.9% 20.1% 0.3% BLI (gray area) was derived by imaging mice without (p/s/cm + ffLuc T cells that were treated with D-luciferin. Data 10 4 Blood

Tcellfluxi ± = – < < Peripheral represent mean SD (n 4 5). **P 0.01, ***P 5152535 0.0% 0.0% 0.0% Days after T cell infusion 0.001; unpaired t test. (B) Longitudinal BLI was per- CD3 formed to monitor NALM-6 tumor cell persistence. Images represent photon flux from NALM-6 cell– B CAR mbIL15-CAR derived rLuc activity. (C) At day 36, tissues and blood Tumor Only + Tumor + Tumor D mbIL15-CARmbIL15-C +Tumor were analyzed by flow cytometry for the presence of + Tumor + + Day CARCAR +Tumor + Tu CD3 T cells and CD19 tumor cells in the three treatment cohorts. Representative flow plots show 22 100 TumorTumor Only On 80 frequencies of CD3 and CD19 in the peripheral blood 500 (n = 3–5), spleen, and BM (n = 4–5). (D) Similar ex- Day 60 al (%) p/s/cm periment was continued to day 98 to assess survival iv 27 (x10 40 (3/7) (n = 7–8 per group). CD4/CD8 ratio of injected cells 20 * 3 2

) (0/8) (0/8) = = /sr Surv was CAR 1:31 and mbIL15-CAR 1:3. Values in 0 Day parentheses represent the fraction of mice surviving 0 20406080100 34 to day 98. *P < 0.05 for mbIL15-CAR versus CAR T-cell Days after tumor challenge 50 treatment, log- (Mantel–Cox).

Recovered T cells exhibited tissue-specific distribution with MRD whereby CAR T cells expressing ffLuc and mbIL15-CAR- reciprocal expression of CD45RO and CCR7, with the highest ffLuc T cells were engrafted in NSG mice, followed 6 d later by + levels observed in BM and peripheral blood, respectively (Fig. i.v. injection of CD19 NALM-6 cells. Only mbIL15-CAR T cells 4C). T-cell differentiation in the peripheral blood and spleen persisted throughout the experiment. In vivo persistence of recapitulated our in vitro observations of the preferential persis- mbIL15-CAR T cells leveled off 19 d after administration, with + tence of CD45ROnegCCR7 T cells, whereas the BM contained stable BM engraftment (Fig. 5 A and C). The mbIL15-CAR T + the highest proportion of more differentiated CD45RO CCR7neg cells displayed potent antitumor activity (Fig. 5B), with no tumor + T cells (Fig. 4D). Similar to in vitro observations, T cells in the (<0.5% of CD19 cells) detected by flow cytometry in the BM, blood expressed moderate levels of CAR and reduced levels of spleen, or peripheral blood (Fig. 5C). In another experiment, cell-surface mbIL15 (SI Appendix,Fig.S6B and C). After ex mice receiving mbIL15-CAR T cells had improved survival: 43% vivo numeric expansion, T cells from these tissues retained their at 98 d compared with 100% mortality in all other treatments by antigen-specific function and produced IFN-γ in response to day 34 (Fig. 5D). These data indicate that mbIL15-CAR T cells + CD19 targets (SI Appendix, Fig. S6D). Together, these data were capable of engrafting in the absence of CD19 antigen or suggest that mbIL15-CAR T cells persist long term in vivo, with a exogenous cytokine, mounting an antitumor response in a low substantial subset exhibiting a clinically desirable TSCM pheno- TAA environment and persisting after tumor clearance, thereby type and continuing to demonstrate specificity for CD19 TAA. making these cells more therapeutically effective in mice than We also evaluated the safety of long-term mbIL15-CAR T-cell conventional CAR T cells. persistence, focusing on potential for transformation or autono- mous proliferation. The mbIL15-CAR T cells were maintained in MbIL15-CAR T Cells Reject Established Leukemia and Sustain continuous in vitro culture for up to 2 y in the absence of antigen Persistence After Clearance of Tumor. Recently, it was reported restimulation and exogenous cytokine. These T cells did not dem- that the IgG4-Fc extracellular domain within CAR structures, onstrate aberrant unrestricted proliferation, and all samples tested such as CD19RCD28 in our initial clinical trials, may bind to (n = 4) maintained a normal karyotype as well as polyclonal Vα/Vβ innate Fcγ receptors (FcγRs), leading to deleterious clearance repertoires (SI Appendix,Fig.S7). In vivo studies of mice with long- and reduced efficacy of infused CAR T cells (36). We hypoth- term (125 d after T-cell infusion) engraftment of mbIL15-CAR T esized that mbIL15 can continue to support CAR T-cell persis- cells in the absence of tumor demonstrated no host toxicity and no tence and act additively with an improved CAR design to evidence of aberrant proliferation of infused genetically modified T augment antitumor efficacy further. Another CAR was engi- cells. Indeed, splenomegaly was not observed, and <1% of human neered (referred as *CAR) with a mutation to eliminate FcγR + ± ± CD3 cells were observed at day 125 (SI Appendix,Fig.S8). binding. We reassessed mbIL15 -CAR and mbIL15 -*CAR T-cell persistence and antitumor efficacy using a model of estab- + MbIL15-CAR T Cells Prevent Leukemia Engraftment and Sustain lished leukemia. NSG mice were injected with ffLuc NALM-6, Persistence After Clearance of Tumor. We assessed survival of followed 6 d later by genetically modified T cells (not modified mbIL15-CAR T cells in a model of minimal residual disease with ffLuc). The *CAR T cells demonstrated greater antitumor (MRD or low TAA burden), where adoptively transferred T cells activity than did CAR T cells, significantly increasing survival may not immediately encounter and be stimulated by TAA, such (P = 0.0007; Fig. 6 A and B). The *CAR T-cell persistence at the as CD19. This assessment was achieved using a model to mimic experimental end point was low (Fig. 6C). Conversely, the

6of10 | www.pnas.org/cgi/doi/10.1073/pnas.1610544113 Hurton et al. Downloaded by guest on October 1, 2021 PNAS PLUS A mbIL15-CARmbIL15-CAR + Tumor+ Tumo mbIL15-*CARmbIL15-*CAR + Tumor+ Tum B 100 )

TumorTumor Only Only CARCAR + +Tumor Tumor *CAR*CAR + +Tumor Tumor (% 10 11 80 P = 0.001 /sr) 10 2 10 60 10 9 40 P = 0.023 10 8 P = 0.007 10 7 20

6

10 Disease-free survival 0 5 Tumor Flux (p/s/cm 10 0 1020304050 5152535455 152535455 15253545 Days after tumor injection Days after tumor injection Tumor Only C CAR mbIL15-CAR *CAR mbIL15-*CAR CAR + Tumor Tumor Only + Tumor + Tumor + Tumor + Tumor mbIL15-CAR + Tumor 96.7% 97.7% 98.4% 85.2% 0.0% *CAR + Tumor mbIL15-*CAR + Tumor Bone Marrow

0.0% 0.0% 0.1% 0.2% 32.2% 83.4% 51.5% 66.9% 84.7% 0.0%

Spleen

GFP (NALM-6) GFP 0.0% 0.0% 0.1% 0.3% 90.0% 16.4% 20.9 19.1% 0.0% INFLAMMATION IMMUNOLOGY AND Peripheral ND Blood

0.0% 1.0% 0.3% 82.4% CD3

Fig. 6. MbIL15-CAR T cells potently reject established systemically disseminated leukemia and persist beyond tumor clearance. NSG mice (n = 7, tumor only group; n = 6, all other T-cell treatments) were injected i.v. with 1.5 × 104 NALM-6 coexpressing GFP and ffLuc. After 6 d of engraftment, 1 × 107 genetically modified T cells were given via intracardiac injection. Infused cells were CD19-specific CAR T cells (expressing CAR or *CAR) with or without coexpression of mbIL15, and infused cell number was based + on CAR T cells. The CD4/CD8 ratio of injected cells was CAR = 1:3, mbIL15-CAR = 1:65, *CAR = 1:25, and mbIL15-*CAR = 1:48. (A) Quantified tumor burden (ffLuc activity) was measured by BLI. Each line represents an individual animal. (B)Kaplan–Meier survival curves show tumor-free survival whereby mice were considered tumor-free if tumor flux was below background. Significance was determined by log rank (Mantel–Cox). (C) Representative flow plots of blood and tissue samples showing frequency of GFP+ (tumor) and human CD3+ (genetically modified T cells) cells harvested from mice (left to right: n = 3, n = 1, n = 2, n = 2, and n = 4).

mbIL15-*CAR T cells demonstrated superior antitumor activity, thereby making these cells more therapeutically effective in mice reducing tumor burden to background levels, with tumor flux irrespective of the potency of the CAR. significantly lower than in mice treated with *CAR T cells (P = 0.006, two-tailed Student’s t test, day 26 after T-cell adminis- Discussion tration). The *CAR T cells induced only temporary tumor re- Adoptive transfer of CD19-specific CAR T cells has demon- gression, followed by aggressive relapse (Fig. 6A). Notably, 83% strated significant antitumor effects as an investigational treat- of mice treated with mbIL15-*CAR T cells exhibited tumor ment for and lymphomas, in which therapeutic elimination, because no responders relapsed during the experi- potential appears to correlate with sustained in vivo persistence ment and flow cytometry corroborated that these mice were (11, 38–41). Moreover, elevated IL15 expression in the tumor tumor-free in the BM, spleen, and peripheral blood (Fig. 6 A and microenvironment correlates with improved patient survival C). Some mice in this group began dying 28 d after T-cell in- (19). We demonstrated that, in contrast to T cells solely jection with no tumor measured by bioluminescence and flow expressing a second-generation CD19-specific CAR, coex- cytometry. Likely, these deaths were from xenograft-versus-host- pression of CAR with a tethered variant of IL-15 augmented disease, as noted by others using cytokine-producing CAR T cells T-cell survival in vitro and in vivo that resulted in superior an- (37). To compensate, we report disease-free survival, rather than titumor effects and was associated with the outgrowth of TSCM. overall survival, which was significantly greater for mice treated Thus, it appears that mbIL15 mimics the transpresentation of with mbIL15-*CAR T cells compared with *CAR T cells (P = endogenous IL-15, signaling via STAT5, without leading to 0.007; Fig. 6B). In vivo persistence of genetically modified T cells aberrant proliferation. was only slightly enhanced by the substitution of CAR with Second-generation CARs are designed with modular cos- *CAR. Substantial improvement was made when mbIL15 was timulatory endodomains to coordinate two signals to mediate coexpressed with *CAR, whereby these mice did not bear tumor CAR-dependent T-cell activation (42), but this design may not at the time of sampling (Fig. 6C). These data indicate that be sufficient in all cases. When TAA is abundant, as when mbIL15-*CAR T cells were capable of engrafting, mounting an treating relapsed B-cell malignancies, the CAR design may be antitumor response in an environment with abundant TAA, and sufficient to mitigate malignant disease. Striking results have persisting after tumor clearance with no evidence of relapse, been observed in several second-generation CAR T-cell clinical

Hurton et al. PNAS Early Edition | 7of10 Downloaded by guest on October 1, 2021 trials that treated acute lymphoblastic leukemia, where complete the early memory phenotype we observed. Thus, it appears that responses were observed in 70–90% of patients (43–45). How- mbIL15 provides signaling to CAR T cells that delivers an ad- ever, long-term results were not as durable, because the overall vantage over other approaches to STAT5-mediated activation, survival was reported to be 65–78% at 6–12 mo (43–45). In possibly due to its ability to generate TSCM. parallel, our mice with established leukemia treated with *CAR One concern for extensively ex vivo-propagated T cells is the T cells showed 50% initial remission but 100% eventually re- potential for generating an exhausted product with limited an- lapsed. This is in contrast to no mice that relapsed following titumor efficacy. Unlike the protective effects of chimeric 4-1BB infusion of mbIL15-*CAR T cells (Fig. 6). This difference sug- costimulation, chimeric CD28 apparently exacerbates exhaustion gests that an mbIL15-mediated signal 3 can bolster the effects induced by chronic CAR signaling, whether by high tumor loads, of the more potent CAR designs, and may augment the T-cell ex vivo numeric expansion, or antigen-independent mechanisms response by inhibiting AICD (16). Alternatively, mbIL15 may (54). Exhausted CAR T cells express higher levels of inhibitory favor the persistence of T cells with a less differentiated state receptors (e.g., PD-1, TIM-3, LAG-3) and increased exhaustion- throughout the multiple activation events of an antitumor re- associated transcription factors (e.g., T-bet, Blimp-1). Such cells sponse. Conversely, when TAA is limited, such as with MRD, exhibit poor production of IL-2 and IFN-γ, increased expression then activation solely through the CAR may be insufficient to of CD25, and decreased levels of CD27 and CD127 (54). The maintain persistence (41). Indeed, in our model with a low leu- AWD-mbIL15-CAR T cells possessed reciprocal features: re- kemia burden, the survival of infused T cells and mice depended duced LAG3 and CTLA4 transcripts (SI Appendix, Fig. S4); low on coexpression of mbIL15 with CAR. The mbIL15-CAR T cells T-BET, BLIMP-1, and IL2RA expression; and increased CD27 persisted in the absence of CD19 TAA, presumably due to sus- and CD127 expression (Fig. 2). The AWD-mbIL15-CAR T cells tained STAT5 phosphorylation. These observations agree with also possessed antigen-dependent IFN-γ production (SI Appendix, studies where engraftment was enhanced by T cells engineered Fig. S3), enhanced IL-2 production (Fig. 3), and antigen-in- to (i) express constitutively active STAT5 (46); (ii) express a dependent persistence (SI Appendix,Fig.S3). Our data indicate that chimeric cytokine receptor mediating constitutive signaling mbIL15 maintains a memory subset of CAR T cells consistent with through CD122 (47); (iii) secrete monomeric recombinant IL-15 TSCM that also preserve their CAR-mediated function. Thus, al- (37, 48); and (iv) express miR155, which increases AKT and though chimeric CD28 costimulation participates in T-cell activa- STAT5 signaling (49). Thus, it appears that coordinated signal- tion, differentiation, and exhaustion, it appears that IL-15 signaling ing with mbIL15 and CAR provides a pathway to improving may block these downstream effects. Additional experiments will be antitumor effects across a spectrum of tumor burden. The ap- needed to understand this interaction and the possible effect of proach of combining signals 1 and 2 via CAR with signal 3 from mbIL15 on activating CAR T cells via chimeric CD137. mbIL15 may be preferred to coinfusing CAR T cells with soluble The cytokines IL-7 and IL-15 appear integral to generating + + recombinant IL-15 due to toxicity (50). CD45RA CCR7 early memory populations (7). We demon- Differentiation status is a predictor of T-cell persistence with strated that mbIL15, and not IL-2 or soluble IL-15 complex, + therapeutic preference given to a biological product that has helps preserve the CCR7 subset (SI Appendix, Fig. S5). Dif- not entered into terminal differentiation or replicative senes- ferent signaling patterns are mediated between soluble and cence (8). It is not clear whether adoptively transferred CAR T transpresented IL-15 (25); thus, the surface expression level and cells modified to receive additional costimulation (e.g., via specific signaling induced by the tethered mbIL15 may provide + constitutively active pSTAT5 or secreted cytokines) can mod- necessary and sufficient signaling to maintain the CCR7 T-cell ulate their activation to form or maintain memory, or if they subset. This finding contrasts with a report that both soluble and remain perpetually supraphysiologically activated, which may membrane-bound IL-15Ra complexes with IL-15 (but not solu- + raise concerns for long-term safety. We observed that mbIL15- ble monomeric IL-15) supported expansion of CD62L and + CAR T cells harvested from TAA-free mice after 50 d persisted CCR7 central memory antigen-specific T cells using an in vitro in a spectrum of differentiation states, including a substantial cell-based cytokine production system (55). Soluble cytokines population exhibiting immature differentiation, delineated as present at supraphysiological levels may induce T-cell differenti- neg + CD45RO CCR7 and consistent with the TSCM phenotype ation (56), which may have been a factor contributing to the ob- (Fig. 4). The in vivo observations recapitulated the in vitro data servations within our in vitro system. Nonetheless, although from long-term cultured AWD mbIL15-CAR T cells (Fig. 3) mechanisms for generating less differentiated T cells remain to be + that were CD45RA and coexpressed CD27 and IL-7Rα,had elucidated fully, we show that these subsets can arise in our system elevated TCF-1, and were capable of IL-2 production (Fig. 2). from signal 3 transmitted by mbIL15. This preserved persistence Some transcription factors function in pairs, with graded re- potential appears analogous to early memory, and, as demon- ciprocal activity controlling effector and memory T-cell differ- strated by others, the least differentiated subsets of adoptively entiation. The ratio of TBET and EOMES is one such pairing, transferred T cells exhibit the greatest therapeutic potential (7, 8). + neg + + with the ratio lowest in early memory T-cell differentiation states Bona fide TSCM (CD45RA CD45RO CD62L CCR7 IL- + + + (33, 51), a pattern observed in mbIL15-CAR and AWD-mbIL15- 7Rα CD122 CD95 ) have been generated in vitro from CAR T cells. Long-lived CAR T cells genetically modified to se- TN cells using a glycogen synthase kinase 3β-inhibitor to activate crete monomeric IL-15 had increased expression of CD45RA, and the WNT pathway, combined with IL-2 (8) or IL-7 and IL-21 (9) + + + a small subset of CD45RA CCR7 cells (∼6%) persisted in the cytokine supplementation. TSCM, albeit CD45RO , have also spleen after an antitumor response (37). We observed that ∼25% been produced using IL-7 and IL-15 ex vivo (7). Our TSCM + + of engrafted splenic T cells were CD45ROnegCCR7 (Fig. 4). In population of mbIL15-CAR T cells (CD45RA CD45ROneg + + + contrast to our results, antigen-specific T cells modified to express CCR7 CD122 CD95 ) had exposure to IL-21– and IL-15–me- constitutively active STAT5 demonstrated an effector phenotype diated signaling during ex vivo numeric expansion, and then only to 20 d after infusion characterized by elevated T-BET levels, ef- IL-15 (in the form of mbIL15) signaling, during the 65-d antigen fector molecule expression, and diminished IL-2 production (52). and cytokine withdrawal assay to generate AWD-mbIL15-CAR T However, another study that used transgenic (Tg) mice expressing cells. Furthermore, TSCM recovered from mice apparently only had a constitutively activated form of STAT5 in the T-lymphocyte access to mbIL15. Although it cannot be ruled out that IL-21 could + compartment showed an increased number of CD8 T cells have a programming effect on the cells, it was primarily upon re- phenotypically similar to memory-like or homeostatically pro- moval of IL-21 and antigen (presented by AaPCs) that we observed liferating T cells (53). This finding suggests that mbIL15-medi- the promotion and outgrowth of the mbIL15-CAR TSCM subset. ated activation of the STAT5 pathway enables the emergence of Previously, it has been shown that IL-15 alone could produce a

8of10 | www.pnas.org/cgi/doi/10.1073/pnas.1610544113 Hurton et al. Downloaded by guest on October 1, 2021 PNAS PLUS small population of TSCM, but combination with IL-7 improved reported that excessive IL-15 can cause malignant trans- these numbers (7, 57). Thus, we contend that the desired TSCM formation of CD4 T cells via inhibition of a negative autor- subset can arise solely upon IL-15–mediated signaling as induced egulatory loop (63). The common thread among these studies is by mbIL15. T-cell supraphysiological activation from elevated levels of IL15- In contrast to observations of IL-7– and IL-15–generated Rα and/or soluble IL-15. However, we provide an alternative to TSCM with no apparent up-regulation of the WNT pathway (7), IL-15 hyperstimulation. We demonstrated that mbIL15-CAR T we noted increased expression of genes in this pathway (TCF7, cells had submaximal pSTAT5 levels relative to control CAR T LEF1, and AXIN2; SI Appendix, Fig. S4), which are regulated by cells stimulated with exogenous IL-15 and that these cells Akt (51, 58). Moreover, because AXIN2 is a direct target of the attained a desired less differentiated phenotype. In other studies, WNT pathway through TCF-1 and LEF-1 activity, its up-regu- T cells constitutively expressing activated STAT5 demonstrated lation suggests that WNT signaling is active in AWD-mbIL15- enhanced persistence but not transformation (46), despite an CAR T cells. Furthermore, Akt signal strength regulates the activated effector phenotype (46, 52). In summary, contrasting WNT pathway, where weaker signaling sustains a memory phe- the previous studies (62, 63, 65, 67), our assessments of AWD- notype (51), suggesting that molecular pharmacology may be mbIL15-CAR T cells did not demonstrate (i) log growth in vitro, beneficial when using exogenous cytokines or priming conditions (ii) uncontrolled expansion in vivo, (iii) aneuploidy, (iv) a ho- that can strongly activate Akt (8, 9, 59). Moreover, our data mogeneous differentiated or activated population, or (v)clonal demonstrated the promotion of an mbIL15-CAR TSCM-like outgrowth, all of which predict favorable human application subset, suggesting that mbIL15 only weakly activates Akt. of mbIL15-CAR T cells. It is significant that our study dem- Our study has benefited from an approach to genetically onstrated long-term engraftment of T cells without dysregulated modifying a population of T cells, which has appeal for the proliferation or activation and no phenotypic, functional, or manufacture of a clinical grade mbIL15-CAR product. In our chromosomal anomalies. antigen and cytokine withdrawal assay, we observed a decline in + + In summary, we demonstrate that CAR T cells can be engi- CD45RO CCR7 T-cell numbers with a concomitant increase of neg + neered to coexpress a tethered form of IL-15 that supports in vivo the CD45RO CCR7 TSCM pool. This finding may be due to cell + + persistence and maintenance of an immature state of differentia- death in the CD45RO CCR7 subpopulation with proliferation in neg + tion through in vitro culture and in vivo engraftment. The signal 3 the CD45RO CCR7 subset. However, this process appears neg + unlikely, considering reports that IL-15 is critical to supporting provided by mbIL15 alone gave rise to CD45RO CCR7 TSCM despite multiple rounds of ex vivo T-cell activation. Our observa- TCM cells (60). Alternatively, it has been shown that a proportion INFLAMMATION neg + tions suggest that mbIL15 may provide a desired signaling enabling IMMUNOLOGY AND of sorted TCM cells dedifferentiated to a CD45RO CCR7 phenotype in the presence of IL-7 and IL-15 and in the absence of infused T cells to prolong persistence, thereby providing durable antigen (57). At this time, we find dedifferentiation to be a plau- immune surveillance and therapeutic potential. sible and likely predominant mechanism generating the mbIL15- Methods CAR TSCM-like cells rather than preferential propagation of small Animal studies were performed in accordance with guidelines set forth by the numbers of TSCM in the genetically modified founder population. Antigen-experienced T cells can drive precocious differentia- M. D. Anderson Cancer Center Institutional Animal Care and Use Committee tion of T cells within mixed populations that impairs antitumor (A3343-01, protocol 03-06-04333). Human mononuclear cell buffy coats from N anonymized healthy donors were obtained from a US Food and Drug Ad- immunity (61). In our system, we modify, ex vivo expand, and ministration-approved vendor or blood bank (Key Biologics or the Gulf Coast infuse cell products containing a mixed grouping of memory Regional Blood Center). Cell lines and their propagation, genetic modifica- T-cell subsets. The heterogeneous mbIL15-CAR T-cell cultures tion of T cells using SB, immunophenotyping, chromium release assay, and BLI show the emergence of TSCM both in vitro and in vivo when were previously described (28, 68–71). Detailed information about experi- removed from antigen. This preserved potential may have clin- mental procedures, antibodies, mice, and statistical analyses can be found in ical appeal, because the least differentiated T-cell subset medi- SI Appendix, Materials and Methods. ated the greatest therapeutic potential (7, 8). It is currently not feasible to treat patients with naturally occurring TSCM due to ACKNOWLEDGMENTS. We thank Dr. Carl June (University of Pennsylvania) their low circulating frequency; thus, identification of strategies for helping to generate the K562-derived AaPC clone 9 cell line and Dr. Perry Hackett (University of Minnesota) for his help with the SB system. Our to generate, or preserve the potential of, clinically relevant appreciation goes to Dr. George McNamara (M. D. Anderson Cancer Center) numbers of genetically modified TSCM would likely contribute to and Dr. Judy Moyes (M. D. Anderson Cancer Center) for editing. Based on the advancement of CAR T-cell–based therapies. The results this work, L.V.H. was a recipient of the Andrew Sowell-Wade Huggins presented herein show that, in our system, mbIL15 promoted a Scholarship in Cancer Research (Cancer Answers Foundation) and Center for T -like population of CAR-modified T cells, and thus bolster Clinical and Translational Sciences T32 Grant Trainee support (NIH). This SCM work was supported by funding from the Center for Clinical and Trans- the rationale for using mbIL15 in further developing clinically lational Sciences, which is funded by the National Center for Advancing compliant strategies to augment this desirable T-cell subset in Translational Sciences of the NIH under Award TL1TR000369; Cancer Center immunotherapy infusion products. Core Grant (CA16672); Research Project Grants R01 (CA124782, CA120956, Supporting T-cell longevity via IL-15 may have risk, because and CA141303); Research Program Project P01 (CA148600); Specialized chronic high exposure to IL-15 could cause aberrant T-cell Programs of Research Excellence Grants (CA100632, CA136411, and CA00632); Albert J. Ward Foundation; Alex’s Lemonade Stand Founda- proliferation or toxicities. In humans, dysregulated IL-15 pro- tion; Burroughs Wellcome Fund; Cancer Prevention and Research Insti- duction, elevated serum levels, or abnormal IL-15 signaling has tute of Texas; Charles B. Goddard Foundation of Texas; CLL Global been associated with autoimmune disease and may be involved in Research Foundation; Energy Transfer Partners; Estate of Noelan L. the pathogenesis of large granular lymphocytic leukemia (62) Bibler; Gillson Longenbaugh Foundation; Harry T. Mangurian, Jr., Fund for Leukemia Immunotherapy; Khalifa Bin Zayed Al Nahyan Foundation; and cutaneous T-cell lymphoma (63). This concern is height- Kleberg Foundation; Leukemia and Lymphoma Society; Lymphoma Research ened, given that Tg mice engineered to overexpress IL-15 de- Foundation; Miller Foundation; Mr. Herb Simons; Mr. and Mrs. Joe H. Scales; veloped leukemia (64, 65); another form of mbIL15 led to Mr. Thomas Scott; National Foundation for Cancer Research; Pediatric Can- autonomous growth of NK cells (66); a T-cell clone modified to cer Research Foundation; Sheikh Khalifa Bin Zayed Al Nahyan Institute for secrete IL-15 led to aberrant autonomous T-cell growth (67); Personalized Cancer Therapy; University of Texas M. D. Anderson Cancer α α Center Sister Institution Network Fund and Moon Shot Fund; and William and in IL-15R -IL-15 double-Tg mice, cis IL-15R expression Lawrence and Blanche Hughes Children’s Foundation. The content of this on leukemic CD8 T cells was required for the development of paper is solely the responsibility of the authors and does not necessarily uncontrolled lymphoproliferation (65). Additionally, it has been represent the official views of the NIH.

Hurton et al. PNAS Early Edition | 9of10 Downloaded by guest on October 1, 2021 1. Eshhar Z, Waks T, Gross G, Schindler DG (1993) Specific activation and targeting of 38. Kalos M, et al. (2011) T cells with chimeric antigen receptors have potent antitumor effects cytotoxic lymphocytes through chimeric single chains consisting of antibody-binding and can establish memory in patients with advanced leukemia. Sci Transl Med 3(95):95ra73. domains and the gamma or zeta subunits of the immunoglobulin and T-cell receptors. 39. Robbins PF, et al. (2004) Cutting edge: Persistence of transferred lymphocyte clono- Proc Natl Acad Sci USA 90(2):720–724. types correlates with cancer regression in patients receiving cell transfer therapy. 2. Savoldo B, et al. (2011) CD28 costimulation improves expansion and persistence of chimeric J Immunol 173(12):7125–7130. antigen receptor-modified T cells in lymphoma patients. J Clin Invest 121(5):1822–1826. 40. Yee C, et al. (2002) Adoptive T cell therapy using antigen-specific CD8+ T cell clones for 3. Porter DL, et al. (2015) Chimeric antigen receptor T cells persist and induce sustained re- the treatment of patients with metastatic melanoma: In vivo persistence, migration, and missions in relapsed refractory chronic lymphocytic leukemia. Sci Transl Med 7(303):303ra139. antitumor effect of transferred T cells. Proc Natl Acad Sci USA 99(25):16168–16173. 4. Gattinoni L, Klebanoff CA, Restifo NP (2012) Paths to stemness: Building the ultimate 41. Brentjens RJ, et al. (2013) CD19-targeted T cells rapidly induce molecular remissions in adults antitumour T cell. Nat Rev Cancer 12(10):671–684. with chemotherapy-refractory acute lymphoblastic leukemia. SciTranslMed5(177):177ra38. 5. Berger C, et al. (2008) Adoptive transfer of effector CD8+ T cells derived from central 42. Kowolik CM, et al. (2006) CD28 costimulation provided through a CD19-specific chi- memory cells establishes persistent T cell memory in primates. JClinInvest118(1):294–305. meric antigen receptor enhances in vivo persistence and antitumor efficacy of 6. Hinrichs CS, et al. (2011) Human effector CD8+ T cells derived from naive rather than adoptively transferred T cells. Cancer Res 66(22):10995–11004. memory subsets possess superior traits for adoptive immunotherapy. Blood 117(3):808–814. 43. Maude SL, et al. (2014) Chimeric antigen receptor T cells for sustained remissions in 7. Cieri N, et al. (2013) IL-7 and IL-15 instruct the generation of human memory stem T leukemia. N Engl J Med 371(16):1507–1517. cells from naive precursors. Blood 121(4):573–584. 44. Lee DW, et al. (2015) T cells expressing CD19 chimeric antigen receptors for acute 8. Gattinoni L, et al. (2011) A human memory T cell subset with stem cell-like properties. lymphoblastic leukaemia in children and young adults: A phase 1 dose-escalation Nat Med 17(10):1290–1297. trial. Lancet 385(9967):517–528. 9. Sabatino M, et al. (2016) Generation of clinical-grade CD19-specific CAR-modified CD8+ 45. Park JH, et al. (2015) Implications of minimal residual disease negative complete re- memory stem cells for the treatment of human B-cell malignancies. Blood 128(4):519–528. mission (MRD-CR) and allogeneic stem cell transplant on safety and clinical outcome 10. Lugli E, et al. (2013) Superior T memory stem cell persistence supports long-lived T cell of CD19 targeted 19-28z CAR modified T cells in adult patients with relapsed, re- memory. J Clin Invest 123(2):594–599. fractory B-cell ALL. Blood 126(23):682. 11. Xu Y, et al. (2014) Closely related T-memory stem cells correlate with in vivo expansion of 46. Grange M, et al. (2012) Activated STAT5 promotes long-lived cytotoxic CD8+ T cells CAR.CD19-T cells and are preserved by IL-7 and IL-15. Blood 123(24):3750–3759. that induce regression of autochthonous melanoma. Cancer Res 72(1):76–87. 12. Alwan LM, et al. (2014) Comparison of acute toxicity and mortality after two different 47. Hunter MR, et al. (2013) Chimeric γc cytokine receptors confer cytokine independent dosing regimens of high-dose -2 for patients with metastatic melanoma. engraftment of human T lymphocytes. Mol Immunol 56(1-2):1–11. Target Oncol 9(1):63–71. 48. Hoyos V, et al. (2010) Engineering CD19-specific T lymphocytes with interleukin-15 13. Matsuoka K, et al. (2013) Low-dose interleukin-2 therapy restores ho- and a suicide gene to enhance their anti-lymphoma/leukemia effects and safety. meostasis in patients with chronic graft-versus-host disease. SciTranslMed5(179):179ra43. Leukemia 24(6):1160–1170. 14. Pipkin ME, et al. (2010) Interleukin-2 and inflammation induce distinct transcriptional pro- 49. Ji Y, et al. (2015) miR-155 augments CD8+ T-cell antitumor activity in lymphoreplete grams that promote the differentiation of effector cytolytic T cells. Immunity 32(1):79–90. hosts by enhancing responsiveness to homeostatic γc cytokines. Proc Natl Acad Sci 15. Zhang X, Sun S, Hwang I, Tough DF, Sprent J (1998) Potent and selective stimulation USA 112(2):476–481. of memory-phenotype CD8+ T cells in vivo by IL-15. Immunity 8(5):591–599. 50. Conlon KC, et al. (2015) Redistribution, hyperproliferation, activation of natural killer 16. Marks-Konczalik J, et al. (2000) IL-2-induced activation-induced cell death is inhibited cells and CD8 T cells, and cytokine production during first-in-human clinical trial of in IL-15 transgenic mice. Proc Natl Acad Sci USA 97(21):11445–11450. recombinant human interleukin-15 in patients with cancer. J Clin Oncol 33(1):74–82. 17. Klebanoff CA, et al. (2004) IL-15 enhances the in vivo antitumor activity of tumor- 51. Kim EH, et al. (2012) Signal integration by Akt regulates CD8 T cell effector and reactive CD8+ T cells. Proc Natl Acad Sci USA 101(7):1969–1974. memory differentiation. J Immunol 188(9):4305–4314. 18. Teague RM, et al. (2006) Interleukin-15 rescues tolerant CD8+ T cells for use in 52. Grange M, et al. (2013) Active STAT5 regulates T-bet and eomesodermin expression in adoptive immunotherapy of established tumors. Nat Med 12(3):335–341. CD8 T cells and imprints a T-bet-dependent Tc1 program with repressed IL-6/TGF-β1 19. Mlecnik B, et al. (2014) Functional network pipeline reveals genetic determinants signaling. J Immunol 191(7):3712–3724. associated with in situ lymphocyte proliferation and survival of cancer patients. Sci 53. Burchill MA, et al. (2003) Distinct effects of STAT5 activation on CD4+ and CD8+ T cell Transl Med 6(228):228ra37. homeostasis: Development of CD4+CD25+ regulatory T cells versus CD8+ memory T 20. Dadi S, et al. (2016) Cancer immunosurveillance by tissue-resident innate lymphoid cells. J Immunol 171(11):5853–5864. cells and innate-like T cells. Cell 164(3):365–377. 54. Long AH, et al. (2015) 4-1BB costimulation ameliorates T cell exhaustion induced by 21. Cheever MA (2008) Twelve immunotherapy drugs that could cure cancers. Immunol tonic signaling of chimeric antigen receptors. Nat Med 21(6):581–590. Rev 222:357–368. 55. Hasan AN, et al. (2016) Soluble and membrane-bound interleukin (IL)-15 Rα/IL-15 22. Stoklasek TA, Schluns KS, Lefrançois L (2006) Combined IL-15/IL-15Ralpha immuno- complexes mediate proliferation of high-avidity central memory CD8(+) T cells for therapy maximizes IL-15 activity in vivo. J Immunol 177(9):6072–6080. adoptive immunotherapy of cancer and infections. Clin Exp Immunol 186(2):249–265. 23. Guo Y, et al. (2015) IL-15 superagonist-mediated immunotoxicity: Role of NK cells and 56. Castro I, Yu A, Dee MJ, Malek TR (2011) The basis of distinctive IL-2- and IL-15-dependent IFN-γ. J Immunol 195(5):2353–2364. signaling: Weak CD122-dependent signaling favors CD8+ T central-memory cell survival 24. Stonier SW, Ma LJ, Castillo EF, Schluns KS (2008) Dendritic cells drive memory CD8 but not T effector-memory cell development. J Immunol 187(10):5170–5182. T-cell homeostasis via IL-15 transpresentation. Blood 112(12):4546–4554. 57. Geginat J, Lanzavecchia A, Sallusto F (2003) Proliferation and differentiation poten- 25. Sato N, Patel HJ, Waldmann TA, Tagaya Y (2007) The IL-15/IL-15Ralpha on cell surfaces tial of human CD8+ memory T-cell subsets in response to antigen or homeostatic enables sustained IL-15 activity and contributes to the long survival of CD8 memory T cytokines. Blood 101(11):4260–4266. cells. Proc Natl Acad Sci USA 104(2):588–593. 58. Wu D, Pan W (2010) GSK3: A multifaceted kinase in Wnt signaling. Trends Biochem Sci 26. Kermer V, Baum V, Hornig N, Kontermann RE, Müller D (2012) An antibody fusion 35(3):161–168. protein for cancer immunotherapy mimicking IL-15 trans-presentation at the tumor 59. van der Waart AB, et al. (2014) Inhibition of Akt signaling promotes the generation of site. Mol Cancer Ther 11(6):1279–1288. superior tumor-reactive T cells for adoptive immunotherapy. Blood 124(23):3490–3500. 27. Rhode PR, et al. (2016) Comparison of the superagonist complex, ALT-803, to IL15 as 60. Boyman O, Krieg C, Homann D, Sprent J (2012) Homeostatic maintenance of T cells cancer immunotherapeutics in animal models. Cancer Immunol Res 4(1):49–60. and natural killer cells. Cell Mol Life Sci 69(10):1597–1608. 28. Singh H, et al. (2013) Manufacture of clinical-grade CD19-specific T cells stably ex- 61. Klebanoff CA, et al. (2015) Klebanoff CA, et al. (2016) Memory T cell-driven differ- pressing chimeric antigen receptor using Sleeping Beauty system and artificial anti- entiation of naive cells impairs adoptive immunotherapy. J Clin Invest 126(1):318–334. gen presenting cells. PLoS One 8(5):e64138. 62. Mishra A, et al. (2012) Aberrant overexpression of IL-15 initiates large granular 29. Kebriaei P, et al. (2016) Phase I trials using Sleeping Beauty to generate CD19-specific lymphocyte leukemia through chromosomal instability and DNA hypermethylation. CAR T cells. J Clin Invest 126(9):3363–3376. Cancer Cell 22(5):645–655. 30. Mortier E, et al. (2006) Soluble interleukin-15 receptor alpha (IL-15R alpha)-sushi as a 63. Mishra A, et al. (2016) Mechanism, consequences, and therapeutic targeting of ab- selective and potent agonist of IL-15 action through IL-15R beta/gamma. Hyper- normal IL15 signaling in cutaneous T-cell lymphoma. Cancer Discov 6(9):986–1005. agonist IL-15 x IL-15R alpha fusion . J Biol Chem 281(3):1612–1619. 64. Fehniger TA, et al. (2001) Fatal leukemia in transgenic mice follows early 31. Rowley J, Monie A, Hung CF, Wu TC (2008) Inhibition of tumor growth by NK1.1+ cells expansions in natural killer and memory phenotype CD8+ T cells. JExpMed193(2):219–231. and CD8+ T cells activated by IL-15 through receptor beta/common gamma signaling 65. Sato N, et al. (2011) Development of an IL-15-autocrine CD8 T-cell leukemia in IL-15- in trans. J Immunol 181(12):8237–8247. transgenic mice requires the cis expression of IL-15Rα. Blood 117(15):4032–4040. 32. Zeng R, et al. (2005) Synergy of IL-21 and IL-15 in regulating CD8+ T cell expansion 66. Imamura M, et al. (2014) Autonomous growth and increased cytotoxicity of natural and function. J Exp Med 201(1):139–148. killer cells expressing membrane-bound interleukin-15. Blood 124(7):1081–1088. 33. Kaech SM, Cui W (2012) Transcriptional control of effector and memory CD8+ Tcell 67. Hsu C, et al. (2007) Cytokine-independent growth and clonal expansion of a primary differentiation. Nat Rev Immunol 12(11):749–761. human CD8+ T-cell clone following retroviral transduction with the IL-15 gene. Blood 34. Michie CA, McLean A, Alcock C, Beverley PC (1992) Lifespan of human lymphocyte 109(12):5168–5177. subsets defined by CD45 isoforms. Nature 360(6401):264–265. 68. Denman CJ, et al. (2012) Membrane-bound IL-21 promotes sustained ex vivo pro- 35. Sallusto F, Lenig D, Förster R, Lipp M, Lanzavecchia A (1999) Two subsets of memory T lym- liferation of human natural killer cells. PLoS One 7(1):e30264. phocytes with distinct homing potentials and effector functions. Nature 401(6754):708–712. 69. Singh H, et al. (2011) Reprogramming CD19-specific T cells with IL-21 signaling can im- 36. Jonnalagadda M, et al. (2015) Chimeric antigen receptors with mutated IgG4 Fc prove adoptive immunotherapy of B-lineage malignancies. Cancer Res 71(10):3516–3527. spacer avoid fc receptor binding and improve T cell persistence and antitumor effi- 70. Cooper LJ, et al. (2003) T-cell clones can be rendered specific for CD19: Toward the cacy. Mol Ther 23(4):757–768. selective augmentation of the graft-versus-B-lineage leukemia effect. Blood 101(4): 37. Markley JC, Sadelain M (2010) IL-7 and IL-21 are superior to IL-2 and IL-15 in pro- 1637–1644. moting human T cell-mediated rejection of systemic lymphoma in immunodeficient 71. Singh H, et al. (2007) Combining adoptive cellular and immunocytokine therapies to mice. Blood 115(17):3508–3519. improve treatment of B-lineage malignancy. Cancer Res 67(6):2872–2880.

10 of 10 | www.pnas.org/cgi/doi/10.1073/pnas.1610544113 Hurton et al. Downloaded by guest on October 1, 2021